ZooScanNet: plankton images captured with the ZooScan

Plankton was sampled with various nets, from bottom or 500m depth to the surface, in many oceans of the world. Samples were imaged with a ZooScan. The full images were processed with ZooProcess which generated regions of interest (ROIs) around each individual object and a set of associated features measured on the object (see Gorsky et al 2010 for more information). The same objects were re-processed to compute features with the scikit-image toolbox http://scikit-image.org. The 1,451,745 resulting objects were sorted by a limited number of operators, following a common taxonomic guide, into 98 taxa, using the web application EcoTaxa http://ecotaxa.obs-vlfr.fr. For the purpose of training machine learning classifiers, the images in each class were split into training, validation, and test sets, with proportions 70%, 15% and 15%.
The folder ZooScanNet_data.tar contains :
taxa.csv.gz
Table of the classification of each object in the dataset, with columns :
- objid: unique object identifier in EcoTaxa (integer number)
- taxon_level1: taxonomic name corresponding to the level 1 classification
- lineage_level1: taxonomic lineage corresponding to the level 1 classification
- taxon_level2: name of the taxon corresponding to the level 2 classification 
- plankton: if the object is a plankton or not (boolean)
- set: class of the image corresponding to the taxon (train : training, val : validation, or test)
- img_path: local path of the image corresponding to the taxon (of level 1), named according to the object id
features_native.csv.gz
Table of metadata of each object including the different features processed by ZooProcess. All features are computed on the object only, not the background. All area/length measures are in pixels. All grey levels are in encoded in 8 bits (0=black, 255=white). With columns:
- objid: unique object identifier in EcoTaxa (integer number) And 48 features:
- area
- mean
- stddev
- mode
- min/max
- perim.
- width,height 
- major,minor
- circ.
- feret
- intden
- median
- skew,kurt
- %area
- area_exc
- fractal
- skelarea
- slope
- histcum1,2,3
- nb1,2,3
- symetrieh,symetriev
- symetriehc,symetrievc
- convperim,convarea
- fcons
- thickr: 
- esd
- elongation
- range
- centroids
- sr
- perimareaexc
- feretareaexc
- perimferet/perimmajor
- circex
- cdexc
See the “ZooScan” sheet - OBJECT metadata, annotation and measurements - , at https://doi.org/10.5281/zenodo.14704250 for definitions.
features_skimage.csv.gz
Table of morphological features recomputed with skimage.measure.regionprops on the ROIs produced by ZooProcess. See http://scikit-image.org/docs/dev/api/skimage.measure.html#skimage.measure.regionprops for documentation.
inventory.tsv
Tree view of the taxonomy and number of images in each taxon, displayed as text. With columns :
- lineage_level1: taxonomic lineage corresponding to the level 1 classification
- taxon_level1: name of the taxon corresponding to the level 1 classification
- n: number of objects in each taxon class 

      2. Second folder ZooScanNet_imgs.tar contains :
imgs
Directory containing images of each object, named according to the object id objid and sorted in subdirectories according to their taxon.

      3. And :
map.png
Map of the sampling locations, to give an idea of the diversity sampled in this dataset.

本数据集的浮游生物样本采用多种网具采集，采样覆盖全球多片海域，采样深度范围为海底或500米至海面。样本通过ZooScan进行成像。完整图像经ZooProcess处理，可在每个独立目标周围生成感兴趣区域（Regions of Interest，ROIs），并提取该目标的一系列关联特征（详细信息参见Gorsky等人2010年的研究）。研究人员还使用scikit-image工具包（http://scikit-image.org）对同一批目标重新处理以计算特征。最终得到的1,451,745个目标对象，由少量操作人员依据通用分类学指南，通过Web应用EcoTaxa（http://ecotaxa.obs-vlfr.fr）划分为98个分类单元。为训练机器学习分类器，每个分类的图像按70%、15%、15%的比例划分为训练集、验证集与测试集。

ZooScanNet_data.tar压缩包包含以下文件：
1. taxa.csv.gz
数据集内各目标对象的分类表，字段如下：
- "objid"：EcoTaxa中的唯一目标标识符（整数）
- "taxon_level1"：一级分类对应的分类单元名称
- "lineage_level1"：一级分类对应的分类学谱系
- "taxon_level2"：二级分类对应的分类单元名称
- "plankton"：目标是否为浮游生物（布尔值）
- "set"：图像所属的数据集类别（train：训练集，val：验证集，test：测试集）
- "img_path"：对应一级分类目标的图像本地路径，文件名以目标ID命名
2. features_native.csv.gz
各目标对象的元数据表，包含ZooProcess处理得到的各类特征。所有特征仅针对目标对象计算，不包含背景。所有面积/长度度量单位均为像素，所有灰度值以8位编码（0代表黑色，255代表白色）。字段包括：
- "objid"：EcoTaxa中的唯一目标标识符（整数），以及48项特征：
- "area"：面积
- "mean"：平均灰度值
- "stddev"：灰度标准差
- "mode"：灰度众数
- "min/max"：灰度最小值/最大值
- "perim."：周长
- "width,height"：宽度、高度
- "major,minor"：长轴、短轴长度
- "circ."：圆形度
- "feret"：费雷特直径
- "intden"：积分光密度
- "median"：灰度中位数
- "skew,kurt"：偏度、峰度
- "%area"：面积占比
- "area_exc"：实体面积
- "fractal"：分形维数
- "skelarea"：骨架面积
- "slope"：灰度斜率
- "histcum1,2,3"：累积直方图1、2、3
- "nb1,2,3"：区域计数1、2、3
- "symetrieh,symetriev"：水平对称性、垂直对称性
- "symetriehc,symetrievc"：水平中心对称性、垂直中心对称性
- "convperim,convarea"：凸包周长、凸包面积
- "fcons"：紧致性参数
- "thickr"：厚度比率
- "esd"：等效球直径
- "elongation"：伸长率
- "range"：灰度范围
- "centroids"：质心坐标
- "sr"：形状比
- "perimareaexc"：周长与实体面积比
- "feretareaexc"：费雷特直径与实体面积比
- "perimferet/perimmajor"：费雷特直径与长轴比值
- "circex"：修正圆形度
- "cdexc"：凸缺陷参数
详细定义可参见https://doi.org/10.5281/zenodo.14704250中的"ZooScan"工作表——"OBJECT metadata, annotation and measurements"部分。
3. features_skimage.csv.gz
基于ZooProcess生成的ROIs，通过skimage.measure.regionprops重新计算的形态特征表。详细文档参见http://scikit-image.org/docs/dev/api/skimage.measure.html#skimage.measure.regionprops。
4. inventory.tsv
分类学谱系与各分类单元图像数量的树形视图文本文件，字段如下：
- "lineage_level1"：一级分类对应的分类学谱系
- "taxon_level1"：一级分类对应的分类单元名称
- "n"：各分类单元类别的目标对象数量

第二部分压缩包ZooScanNet_imgs.tar包含：
imgs/
存储各目标对象图像的目录，文件名以目标ID（objid）命名，并按照其所属分类单元存入对应子目录。

此外还包含：
map.png
采样位置分布图，用于直观展示本数据集覆盖的采样多样性。

搜集汇总

背景与挑战

背景概述

ZooScanNet是一个大规模的浮游生物图像数据集，包含1995年至2019年间在全球海洋中采集的1,451,745个对象图像，使用ZooScan设备成像并经过人工分类为98个分类群。该数据集专门为机器学习分类任务设计，已分割为训练、验证和测试集，并提供了图像、分类表和形态特征元数据，适用于浮游生物研究和计算机视觉应用。